确定声源频率和位置的超分辨阵列处理方法  

A high resolution array processing for frequencies estimation and location of sources

作　　者：俞靖[1] 杨春亭[2] 王学礼[3] 

机构地区：[1]南京航空航天大学,南京210016 [2]杭州应用技术工程学院,杭州310012 [3]西安交通大学,西安710049

出　　处：《声学学报》1999年第1期45-52,共8页Acta Acustica

摘　　要：基于均匀线阵接收一定距离的声源得到的时间和空间维的样本数据提出分离处理：用时间协方差矩阵确定声源频率；通过对近场空间协方差矩阵进行修正，用重构在相同方位上的远场近似空间协方差矩阵确定声源方位角。并在此基础上，通过改变阵列中心位置确定声源位置。计算机模拟和声学实验证实了上述方法的可行性．A separable approach based on a space-time two dimensional data matrix from the finite distance of sources using a equi-spaced line array is proposed. By using the time covariance matrix,the frequencies of multiple sources can be obtained. By using the reconstructed far-field approximation space covariance matrix at the same bearings from the revised near-field space covariance matrix, the bearings of multiple sources can be obtained. The method can be applied to locate sources by change position of the array center. The simulation and experiment results support the theoretical predications.

关 键 词：超分辨阵列处理 声源频率 声源位置 多声源 

分 类 号：O422[理学—声学]